Firstly the study population was chosen from a single managed health care plan in Colorado, which may limit the generalizability of their results. Secondly, it is possible that physicians are possibly making a analysis of varicella in children who are unvaccinated than in children who are vaccinated. This type of diagnostic prejudice would lead to an misjudge of the risk related with vaccine refusal. However, breakthrough cases in vaccinated patients are common. In their study, 95% of varicella victims were in children who were immunized against varicella. For this reason, they do not trust that a diagnostic prejudice had a large effect on their results. Thirdly, vaccine rejection in their case-control population was unusual; there were only 10 vaccine deniers (1.6%) of the 626 victims and controls. As a result, they did not have enough statistical power to assess the link between vaccine rejection and varicella infection by year from 1998 to